Using gestures as a means for interfacing with portable devices is becoming commonplace. For example, current Android® based phones can control access through sensing and accepting security gestures in place of alphanumeric passwords. Such gestures are typically recognized by analyzing kinematic (motion related) data provided by sensors found in the portable device. Kinematic sensors include accelerometers, gyroscopes, magnetometers, etc.
However, human motions are usually irregular and in 3-D. Many existing methods of identifying 3-D motions require complex computations and integrations of 3-D data from multiple accelerometers or other motion sensors, which not only consume a large amount of CPU time but can lead to a higher margin of error.
In consequence, the present inventors have recognized that there is value and need in providing a method and system for recognizing 3-D motions on 2-D planes, which only require minimized computations and integrations.